AI-driven blood test uses epigenetic fingerprints to flag high-risk prediabetes

A German study shows that an AI-enhanced blood test analyzing DNA methylation markers can classify people into high- or moderate-risk prediabetes groups with about 90% accuracy, using 1,557 epigenetic markers to form a biological fingerprint. This approach could enable earlier, personalized prevention and reduce progression to type 2 diabetes, potentially serving as a simpler, cost-effective diagnostic tool compared with extensive clinical testing.